PROJECT SUMMARY An increasingly prominent theme in working memory (WM) research is the idea that information can be held in WM in different states of priority. The PI's group has explored this with a procedure that unconfounds the contents of WM from the focus of attention. In this dual serial retrocuing (DSR) task, one of two items held in WM is cued as the target of an impending memory probe, thereby acquiring the status of ?attended memory item? (AMI). Although the uncued item (the ?unattended memory item,? UMI) can't be dropped from WM (because it may be tested later in the trial), multivariate evidence for its active retention, as measured with functional magnetic resonance imaging (fMRI) or electroencephalography (EEG), drops to baseline levels. This novel and unexpected finding ? that information in WM can lack a detectable active trace -- has prompted several important questions that will be addressed by pursuing three specific aims: Aim 1: To test the hypothesis that the cognitive control of unattended memory items (UMI) is implemented by the same frontoparietal mechanisms that control spatial and nonspatial attention. Aim 2: To test the hypothesis that the selection of visual stimuli, whether from the environment or from WM, is accomplished, in part, by the hijacking of low-frequency oscillatory dynamics that are fundamental to the waking- state physiology of the corticothalamic circuitry of the visual system. Aim 3: To test the hypothesis that the function of context binding contributes to delay-period activity of the posterior parietal cortex (PPC). These Aims will be pursued with experiments that employ extracranial EEG, in many cases with concurrent transcranial magnetic stimulation (TMS), in neurologically healthy young adults, as well as with experiments acquiring electrocorticography (ECoG) recordings from patients undergoing presurgical planning as a part of the clinical treatment of their epilepsy. From a broad perspective, the research program proposed here addresses a question that is of central importance for cognitive neuroscience and for mental health: What are the mechanisms whereby we control the contents of our thoughts? Its potential implications for mental health derive both from the psychological processes for which it has relevance (e.g., ruminative thought patterns, cognitive reappraisal, hallucination, processing trauma), and from the neurophysiological mechanisms that are disordered in psychiatric disease (e.g., patterns of disordered neural oscillations in schizophrenia).